Seems we just need to use the units of the variable linked with the "climatology" attribute.
Here is an ncgen that demonstrates the problem. climatology_bounds variable will be read in as Float64 rather than DateTime
netcdf climatological_seasons {
dimensions:
time=4;
nv=2;
lat=10; // added
lon=10; // added
variables:
float temperature(time,lat,lon);
temperature:long_name="surface air temperature";
temperature:cell_methods="time: minimum within years time: mean over years";
temperature:units="K";
double time(time);
time:climatology="climatology_bounds";
time:units="days since 1960-1-1";
double climatology_bounds(time,nv);
data: // time coordinates translated to datetime format
time="1960-4-16", "1960-7-16", "1960-10-16", "1961-1-16" ;
climatology_bounds="1960-3-1", "1990-6-1",
"1960-6-1", "1990-9-1",
"1960-9-1", "1990-12-1",
"1960-12-1", "1991-3-1" ;
}
If you save that as climatological_seasons.ncgen and run:
using NCDatasets
using NCDatasets.NetCDF_jll
NetCDF_jll.ncgen() do exe
run(`$exe -k nc4 -b -o clim.nc climatological_seasons.ncgen`)
endThen load the file with NCDataset, you get a Float64 eltype for "climatology_bounds"
julia> NCDataset("clim.nc")
Dataset: clim.nc
Group: /
Dimensions
time = 4
nv = 2
lat = 10
lon = 10
Variables
temperature (10 × 10 × 4)
Datatype: Float32 (Float32)
Dimensions: lon × lat × time
Attributes:
long_name = surface air temperature
cell_methods = time: minimum within years time: mean over years
units = K
time (4)
Datatype: DateTime (Float64)
Dimensions: time
Attributes:
climatology = climatology_bounds
units = days since 1960-1-1
climatology_bounds (2 × 4)
Datatype: Float64 (Float64)
Dimensions: nv × time
Seems we just need to use the units of the variable linked with the "climatology" attribute.
Here is an ncgen that demonstrates the problem.
climatology_boundsvariable will be read in asFloat64rather thanDateTimeIf you save that as climatological_seasons.ncgen and run:
Then load the file with
NCDataset, you get aFloat64eltype for "climatology_bounds"